The data were collected with a web-based survey between January and March 2021 as a part of the “EHR systems as a tool for physicians 2021” study [21]. An invitation to participate was sent by email to all working-age (aged 65 y and younger) physicians from the Finnish Medical Association’s register. A detailed description of the data collection, measures used in the questionnaire and the evolution of the questionnaire can be found elsewhere [21]. Altogether 4683 physicians (estimated response rate 24.5%) responded to the survey. We excluded 53 respondents who indicated that they did not use EHRs at all. Thus, the final sample included 4630 physicians. The sample was representative of the target population, with a slightly more older physicians, specialists, and hospital physicians compared to the eligible population [22].

The study was conducted according to national laws, regulations, and the Declaration of Helsinki. The survey did not require a statement from the ethics committee, given that the Finnish National Board on Research Integrity [23] has defined that the administration of surveys to gather respondents’ opinions, which are not anticipated to result in harm, does not necessitate a statement from the ethics committee. All respondents provided informed consent to participate, participated voluntarily, and received written information about the study. All responses were anonymous and the research data did not include any personally identifiable information. No compensation was provided to participants.

The variables used in this study can be found in Multimedia Appendix 1.

Perceived changes in clinical work due to digitalization were measured by respondents’ opinions on the following statements on how the digitalization of health care had changed their work in the past 3 years:

The statements were rated: 1=fully agree, 2=somewhat agree, 3=neither agree nor disagree, 4=somewhat disagree, and 5=fully disagree, which were then recoded as 0=disagree or neutral (response options 3‐5) and 1=agree (1-2). These statements were developed by clinicians and healthcare digitalization experts according to the strategic goals of digitalization of health care [4] and pilot tested with physicians. The statements have been associated, for example, with work-related time pressure and stress related to information systems [14].

Perceived management support from HISs was assessed with 3 items (Cronbach α=0.81) measuring the supportiveness of information systems for knowledge management. The respondents were directed to this section if their professional title indicated that they might have a leading or managerial position. Furthermore, the respondents were advised that if they do not have administrative or management responsibilities, they should not grade these statements (817 responded to these statements). They were asked how, in all, do the information systems used in their organization work as tools for leadership and management The items were: “I use information systems daily for activity monitoring,” “information systems facilitate monitoring the quality of activities,” and “information systems help me to monitor the targets set by my unit (eg, numbers of patients, periods of treatment, and types of operations).” The response options with 5-point Likert scale ranged from 1=fully disagree to 5=fully agree. A mean of these three items was calculated, higher scores indicating better support for management. Respondents were advised not to answer this question if they do not have administrative or management responsibilities; thus, those who answered this question were considered as being in leadership position.

Furthermore, the survey asked respondents’ age (which was categorized as under 35, 35‐44, 45‐54, and 55‐64 years) and employment sector (hospitals, primary care, private, and other such as universities). Respondents’ gender was received from registers.

First, we examined whether being in a leadership position was associated with perceived changes in work due to digitalization in the total sample. Multivariable logistic regression analyses were conducted with the perceived changes in clinical work due to digitalization (improved preventive work, facilitated access to patient information, progressed interprofessional collaboration, and accelerated clinical encounters in separate analyses) as dependent variables and leadership position (yes or no), gender, age, and sector as independent variables.

Second, we examined whether the variables related to perceived changes in work due to digitalization were associated with perceived management support from HISs among those who had administrative or management responsibilities (n=817). Analyses of covariance were conducted, with perceived management support from HIS as the dependent variable and the four statements regarding perceived changes in clinical work due to digitalization as independent variables (in the same analysis), adjusted for gender, age, and sector. All analyses were conducted using IBM SPSS Statistics software (version 28).

This study found that physician leaders agreed more often that digitalization had improved preventive work, facilitated access to patient information, progressed interprofessional collaboration, and accelerated clinical encounters compared to those without leadership positions. This suggests that leaders have a more positive view related to ramifications of digitalization than physicians without leadership positions. However, physician leaders did not experience that HISs support their management work, given that their average rating for support was 2.7 on a scale of 1 to 5. Nevertheless, although physician leaders considered HISs support for leadership and management insufficient, those who agreed with these statements also gave more positive responses about the impacts of digitalization.

Our finding that physician leaders had a more positive view related to digitalization than other physicians is congruent with a previous study [5]. A qualitative study among health and social care professionals suggests that leaders may overlook the new digital skills needs and problems caused by combining continuous learning with busy work schedules [5]. A possible reason why health care leaders may have a more positive attitude toward digitalization is that they play an active role in leading these initiatives by striving to demonstrate a visible commitment towards new DHTs [24]; they also try to convince reluctant users to envision digitalization in a more favorable manner [25]. It has been suggested that the successful implementation of DHTs among health care professionals is contingent upon managers being convinced of its positive effects [26].

Our study found that those leaders who agreed with the positive statements related to the changes in clinical work due to digitalization also perceived more often that HISs supported management work compared to those who disagreed or were neutral. Due to the cross-sectional nature of our study, we cannot identify the causality of this association; that is, we cannot definitely identify whether effective support from HISs leads to improved efficiency in management work for leaders and contributes to a positive view toward digitalization. Alternatively, it could be that those who have noticed that digitalization has improved preventive work, facilitated access to patient information, promoted collaboration, and accelerated clinical encounters are therefore also more inclined to rate that HISs support their management work. It is possible that the general satisfaction with the benefits of digitalization for leaders in health care is reflected in their responses. Thus, it seems important to ensure that leaders keep themselves well-informed about the realities of clinical work and thereby avoid having excessively positive view related to digitalization [26]. Many physician leaders participate also to clinical work which makes it easier to stay up to date about the situation in clinical care. However, physicians working in nonleadership positions need to communicate work-related issues and digital challenges to management, providing nuanced insights for decision-making [26].

Our results suggest sector-specific variations, with physicians working in specialized care perceiving less improvement in preventive work and facilitated access to information compared to physicians working elsewhere. Whereas primary care focuses on preventive measures and health promotion, specialized care physicians mainly treat the diseases in their own specialty [27]. A previous study among specialized care physicians showed that day-to-day operations management HISs supported decision-making to some extent but did not improve access to information [18]. Hospital physicians have also reported that they need to use a number of information systems to support decision making instead of one HIS which they would prefer [18]. Hospital physicians have also reported that they were more able to monitor the use of personnel, equipment, or room resources through HISs compared to primary care physicians [17]. Furthermore, physicians in specialized care settings have previously stated fragmentation of information and data overload resulting from difficult-to-use EHRs [28]. In addition, the implementation of HISs has led to changes in professionals’ workflow, potentially causing cognitive burden [29]. Notably, high levels of stress related to HISs have been reported, particularly among physicians working in specialized care [12]. These challenges may contribute to the perception of less improvement in access to information among physicians in specialized care. Furthermore, contextual factors such as workload, resource allocation and availability, and on-call work may have affected these sector-specific differences.

Physician leaders in primary care perceived that HISs supported less their management work than leaders from other sectors. This is congruent with previous results in Finland which showed that, of different working sectors, leaders from primary care were most dissatisfied with HISs in guiding daily activity, facilitating measurement of functional quality, and monitoring the achieving of the set targets [17]. Primary care leaders have been found to use less managerial HISs compared to physicians in specialized care and private sector [17]. Compared to other sectors, primary care differs in aspects such as care processes, outcome goals, and the validity of summative quality scorecards [30]; thus, needs for managerial HISs also vary. For example, patient-centered reporting and outcomes (eg, reduction in days of avoidable disability) have been suggested as priorities for quality management in primary care [30].

In our study, older physicians were less likely to perceive that digitalization has improved preventive work or facilitated access to information than younger physicians. This corresponds with previous findings that, compared to younger physicians, older physicians are less likely to perceive that HISs will improve data and care coordination [31]. Older physicians have also been found to perceive less affinity toward digitalization, as well as use and trust mobile apps less in their professional lives than younger physicians [32]. Our finding that older physicians did not consider digitalization as having facilitated preventive work, may influence their likelihood to recommend preventive DHTs to patients. Consequently, this may impact the extent of digital preventive work, as patients’ unawareness of the potential DHTs or their benefits has been mentioned amongst the barriers of digital prevention [33].

Our study did not find many gender differences. However, we found that women were less likely to perceive that digitalization had accelerated clinical encounters compared to men. A previous study among physician students did not find gender differences related to fear of digital challenges, but women rated their knowledge in different fields of digital medicine as worse than men, whereas men frequently reported keeping themselves informed about digital medicine outside of their studies [34]. It has been found that women physicians tend to perceive digitalization-related differences in working conditions as more stress-reducing than men physicians [35].

There are some limitations to this study. A major limitation of our study is the use of cross-sectional survey data, thus we cannot draw any causal inferences. Therefore, according to our data it is not possible to say whether efficient support from HISs to management work leads to more positive views toward digitalization or vice versa. Thus, future studies with longitudinal design should examine more thoroughly the causal associations between HISs support for management and perceptions of digitalization. Furthermore, a question related to problems associated with an inflation of the strengths of relationships and with common method variance always arise when using self-rated measures. Self-reported data are subject to many biases such as recall, desirability, and response biases along with problems in accuracy and generalizability [36]. The reliability of the variable measuring how HISs support management was .81; thus, it can be considered reliable.

In addition, even though we adjusted our analyses for age, gender, and employment sector, we cannot put aside the possibility of residual confounding. Some variables that were not adjusted for, such as competence in knowledge management, may have influenced the associations we examined and, thus, cause bias to our results. Furthermore, usability of HISs and experience as an HIS users are among factors that might have affected our findings, for example, by influencing the overall attitude toward HISs. We excluded 53 respondents who did not use EHRs at all; thus, their opinion was not considered here. This was done because we reasoned that their work did not involve such tasks that they could accurately answer our questions related to digitalization and the supportiveness of information systems for knowledge management.

Our results can be considered generalizable to other countries with universal access to health care and long history of HISs [37]. Furthermore, generalizability of our findings is compromised by rather low response rate of our study, which indicates likelihood of nonresponse bias. However, our sample was rather large and it has been shown to be representative of the target population [22]. The response rates have been found similarly low also in other physician studies [38]. Unfortunately, it seems that response rates among health and social care professionals especially in Finland seem to have decreased. A potential factor contributing to the declining response rates may be survey fatigue, given the high volume of various surveys conducted in this sector in Finland. Another reason for low response rate in our survey may be that the survey was conducted via email, which may have caused problems in message deliveries and email addresses might have changed.

This study found that physician leaders were more optimistic related to the changes in clinical work due to digitalization compared to other physicians. Furthermore, those leaders who perceived these changes positively also perceived that HISs supported their management work. Leaders should comprehensively assess and address the perceptions of physicians in routine clinical work concerning work and its changes. This approach ensures that leaders possess current and accurate information to optimize planning for personnel resources, training initiatives, and future DHT implementations.

Furthermore, to ensure that leaders have a positive view related to digitalization, it is important that HISs support their management work. Thus, it is important to focus on the quality and usability of information systems. For example, it has been suggested that HISs designed for leaders should be flexible and consider the needs of different users to support a shared situational awareness and operational intelligence in the management [39]. It is important that the success of HIS development would be regularly monitored and that future studies would address HISs’ effect to clinical and management work. Especially, because our study was cross-sectional, future studies with longitudinal study design are needed.